It's a type. A datatype. Like int or long or size_t or unsigned char.
You know how in C, size_t is an unsigned integer large enough to represent any array size possible? The C++ std::string class provides string::size_type as an integer datatype large enough to represent any possible string size.
The 'find' member function returns the location of the first occurrence of a substring or character in a string. This return value is an integer (whose type is string::size_type). If the substring or character is not found, it needs some value to return that signals this fact. A library designer could have said, "okay, let's return -1 when the substring's not found." But instead, they return some integer value hardcoded as string::npos. It's probably -1. Code that uses string::npos is more readable than code that uses -1. (Since you can't assume that string::npos == -1, you have to use string::npos anyway.)
Isn't it about time forums rewarded their contributors?
Contribute to this discussion and earn rewards points that can be cashed out for dollars.